Despite the emergence of viruses resistant to a combination of antiviral drugs, a substantial number of patients show consistent improvement of either their viral level, immunological or both parameters. The PI's hypothesis is that the improvement is due to viruses that are less fit as a result of resistance conferring and coinciding mutations that lead to an attenuated phenotype. It has been demonstrated that didanosine (ddI)-induced mutation L74V and 3TC-selected mutation M184V render a replication disadvantage to the virus and the altered RTs are less processive. Comparison of replicative fitness of wild type, L74V and M184V viruses and the use of mathematical modeling showed that both RT variants are less fit than the wild type virus. Notably, L74V mutation in RT has been shown to be selected by several nucleoside RT inhibitors, including ddI, ddC, ABC and DXG. Similarly, certain point mutations selected by protease inhibitors have also been shown to impair HIV replication. While the relative development of drug-selected mutations have been delayed with the use of HAART during clinical trials, several recent studies have shown that MDR escape mutant viruses can arise during therapy. It is not clear, how MDR mutations interact and affect viral replication. The overall novelty of this proposal is to understand how MDR mutations of HIV-1 RT and protease interact and affect HIV fitness. The PI proposes to analyze the fitness of sequential clinical isolates that contain drug-selected mutations in RT and PR genes of the virus. MDR Clinical isolates that correspond to reduced viral load, and/or increased CD4+ cell numbers during the course of therapy will be analyzed for viral fitness. Selected mutations will be inserted in pNL4-3, and in the background of L74V RT and replication kinetics of cloned viruses will be analyzed. The specific aims of this proposal are to: 1) analyze HAART-selected HIV multidrug resistant variants for growth fitness; 2) characterize the interaction of RT and protease inhibitors-selected mutations in cloned viruses; 3) determine the processivity of altered RT, and 4) determine the development of resistance mutations in L74V RT variant. The effect of combination of different mutations on the viral replication will be studied in PBMC culture and relative growth fitness will be calculated with a mathematical model. The insight provided through the proposed studies should help optimize protocols of specific drug combinations for clinical trials, based on drugs which select for viruses that are less fit and design a salvage therapy for multi-drug failure subjects.